1. Field of the Invention
The invention relates to the transmission of signals by optical means and more particularly to high bit rate transmission on long-haul links using optical fibers.
The invention relates to a system for dynamically compensating at least some polarization dispersion observed in fiber optic transmission systems.
2. Description of the Prior Art
A fiber optic transmission system typically includes:
a transmitter terminal using at least one optical carrier wave whose optical frequency and/or power it modulates according to the data to be transmitted,
an optical transmission line consisting of at least monomode fiber section conveying the signal emitted by the transmitter terminal, and
a receiver terminal receiving the optical signal transmitted by the fiber.
The performance of an optical transmission system, in particular in terms of signal quality and bit rate, is limited in particular by the optical properties of the link, which is the site of physical phenomena which degrade the optical signals. Of all the various phenomena that have been identified, attenuation of the optical power and chromatic dispersion initially appeared to be the most severely constraining and means have been proposed for at least partially remedying the degradation that they cause.
The attenuation in fibers of a given type depends on the signal carrier wavelength. Monomode fibers installed during the past decade and referred to as xe2x80x9cstandard fibersxe2x80x9d have a minimum attenuation at a wavelength around 1.5 xcexcm, which shows the benefit of choosing this value for the carriers.
Also, to increase transmission distances further, attenuation has been compensated by means of optical amplifiers at the upstream or downstream end of the link or all along the link.
Chromatic dispersion also depends on wavelength. For standard fibers there is zero chromatic dispersion at 1.3 xcexcm and the chromatic dispersion at 1.5 xcexcm is approximately 1.7 ps/(km.nm). The low attenuation at 1.5 xcexcm has led to the development of new fibers referred to as xe2x80x9cdispersion shifted fibersxe2x80x9d for which there is zero chromatic dispersion at this wavelength.
Nevertheless, to improve the performance of standard fibers already installed, attempts have been made to correct the effects of chromatic dispersion in these fibers at 1.5 xcexcm.
One solution is to insert into the link at least one dispersion compensating fiber (DCF). To compensate the chromatic dispersion exactly, it is sufficient for the dispersion compensating fiber to have a length and dispersion characteristics such that the cumulative dispersion along the compensating fiber is equal and opposite to that created along the fiber of the transmission link.
A residual cumulative dispersion value DR can therefore be defined for the whole of the link, including the compensating fiber(s), as the algebraic sum of the cumulative dispersions DL and DC of the dispersion compensating fiber(s) and the fiber of the transmission link. In mathematical terms this can be expressed by the equation:
DR=DC+DL=∫D1(z1).dz1+∫D2(z2).dz2xe2x80x83xe2x80x83(1)
in which z1 and Z2 are the abscissas of points respectively along the dispersion compensating fiber and along the associated link and D1 and D2 are the respective chromatic dispersion parameters at the abscissas z1 and z2 of the dispersion compensating fiber and of the fiber of the transmission link. The integrals which express the cumulative dispersions DC and DL are respectively calculated along the dispersion compensating fiber and along the associated transmission link fiber, taking the direction of propagation of the waves as the positive direction.
It will be remembered that the dispersion parameter D is related to the propagation constant xcex2 by the equation:
d2xcex2/dxcfx892=xe2x88x92(2xcfx80c/xcfx892)D,
in which xcfx89 is the angular frequency of the wave and c is the speed of light in a vacuum.
The condition for exact compensation of chromatic dispersion is therefore DR=DC+DL=0.
In reality exact compensation of chromatic dispersion is not always the optimum, because the quality of the compensated signal received also depends on other transmission parameters and in particular on how the transmitted signal is modulated. This applies in particular if the transmitted signal includes a xe2x80x9cchirpxe2x80x9d, i.e. optical frequency modulation accompanying any amplitude modulation.
In fact, this kind of compensation is applied only when required, i.e. when transmission conditions (fiber type, modulation type, transmission distance and bit rates) would, if not compensated, lead to bit error rates exceeding a commercially acceptable limiting value, typically 10xe2x88x9215. Moreover, to minimize the cost of the dispersion compensating fiber, a minimum compensation value is normally chosen that is compatible with the required bit error rate. Thus for sufficiently short links no attempt is made to compensate chromatic dispersion.
Until now, the forms of compensation referred to above have been treated independently and without regard to another unfavorable phenomenon referred to as xe2x80x9cpolarization modal dispersionxe2x80x9d. Under current operating conditions for optical transmission systems, this phenomenon has long been considered as negligible compared to chromatic dispersion. This is no longer the case when the length, and most importantly the bit rate, of the links are to be increased further.
Even in the absence of chromatic dispersion in the usual sense, and although the carrier wave supplied by a laser diode in the transmitter is totally polarized, the fibers are the site of polarization dispersion which has the effect, for example, that a pulse emitted by the transmitter terminal is received in a distorted form after propagating in a fiber and has a duration greater than its original duration.
This distortion is due to the birefringence of the fibers, whose effect is that the optical signal is depolarized during transmission. To a first approximation, the signal received at the end of the link fiber can be considered as made up of two orthogonal components, one corresponding to a state of polarization for which the propagation speed is a maximum (fastest principal state of polarization) and the other corresponding to a state of polarization for which the propagation speed is a minimum (slowest principal state of polarization). In other words, the pulsed signal received at the end of the link fiber can be considered to be made up of a first pulsed signal which is polarized with a privileged state of polarization and arrives first and a second pulsed signal propagating with a retarded state of propagation and arriving with a delay referred to as the differential group delay (DGD) which depends in particular on the length of the link fiber. The above two principal states of polarization (PSP) therefore characterize the link.
Consequently, if the transmitter terminal emits an optical signal consisting of a very short pulse, the optical signal received by the receiver terminal is made up of two successive pulses polarized orthogonally and having a relative time difference equal to the differential group delay. As detection by the terminal entails providing an electrical form of the measured total optical power received, the detected pulse has a temporal width increased as a function of the value of the differential group delay.
The delay can be in the order of 50 picoseconds for a standard fiber 100 kilometers long. The distortion of the pulses received by the receiver terminal can cause errors in decoding the transmitted data and polarization dispersion therefore constitutes a factor limiting the performance of optical links, whether analog or digital.
It is currently possible to manufacture monomode fibers with low polarization dispersion (approximately 0.05 ps/km). However, the problem remains in the case of xe2x80x9cstandard fibersxe2x80x9d already installed and which have very high polarization dispersion, constituting a major technical obstacle to increasing the transmitted bit rate. Also, this problem will arise again for low polarization dispersion fibers when it becomes necessary to increase the bit rate further.
It is also possible to manufacture fibers with high polarization dispersion, referred to as polarization maintaining fibers (PMF), which can be used in short sections to procure a fixed differential group delay with principal states of polarization that do not vary. By judiciously disposing a component of this kind (or any device generating a differential delay between two orthogonal modes of polarization) in series with a transmission link which is subject to polarization dispersion it is possible to achieve optical compensation of the polarization dispersion. This can be achieved either by using a polarization maintaining fiber with the same differential delay as the link, but interchanging the fast and slow principal states of polarization, or by causing a principal state of polarization of the combination of the link and the polarization maintaining fiber to coincide with the state of polarization of the source of emission. A polarization controller is used for this purpose and is inserted between the link and the polarization maintaining fiber.
One important aspect of polarization modal dispersion is that the value of the differential group delay and the principal states of polarization of a link vary in time as a function of many factors, such as vibration and temperature. Accordingly, unlike chromatic dispersion, polarization dispersion must be considered to be a random phenomenon. In particular, the polarization dispersion of a link is characterized by a polarization mode dispersion delay (PMD) defined as the average measured value of the differential group delay.
To be more precise, it can be shown that the polarization dispersion can be represented by a random rotation vector xcexa9 in the Poincarxc3xa9 space in which the states of polarization are usually represented by a state of polarization vector S, referred to as the Stokes vector, whose end is situated on a sphere. FIG. 1 shows the principal vectors concerned: the state of polarization vector 5, the polarization dispersion vector xcexa9 and the vector of the principal states of polarization e, "PHgr" being the angle between S and xcexa9.
The vectors e and xcexa9 are in the same direction and the following equation applies: ∂S/∂xcfx89=xcexa9{circle around (xc3x97)}S, where xcfx89 is the angular frequency of the optical wave and the symbol {circle around (xc3x97)} designates the vector product.
The modulus of xcexa9 is the value of the group time difference, i.e. of the propagation delay between two waves polarized with the two main states of polarization of the link.
Another consequence of the random nature of the phenomenon is that a compensating system must be adaptive and the differential group delay of the polarization maintaining fiber must be made at least equal to the differential group delay values to be compensated. One such compensating system is described in European patent application EP-A-853 395 filed Dec. 30, 1997 and published Jul. 15, 1998.
A problem which has come to light in work on compensating PMD is that of the combined effects of polarization dispersion and chromatic dispersion. It has been found that in reality the efficacy of PMD compensation is highly sensitive to the residual chromatic dispersion value Dr of the link as a whole and therefore to the existence and the chosen value of chromatic dispersion compensation.
In particular, it has been found necessary to introduce precise chromatic dispersion compensation even for links where such compensation would not be necessary in the absence of PMD.
It has also been found that the optimum value of chromatic dispersion compensation to be applied in the presence of PMD does not always correspond to the optimum value of compensation that would be applied in the absence of PMD, for example in the case where the signal transmitted includes a xe2x80x9cchirpxe2x80x9d.
This phenomenon can be explained by the fact that the method used to compensate PMD effects a first order compensation relative to the angular frequency xcfx89. In reality, the polarization dispersion vector xcexa9 and the principal states of polarization depend on the angular frequency xcfx89 and this dependency becomes dominant when first order compensation is effected. The second order effects in respect of xcfx89 can in part be treated as additional chromatic dispersion induced by PMD. As PMD is random, this induced chromatic dispersion will in turn be random.
Accordingly, the object of the invention is to improve the efficacy of polarization compensation by allowing for the foregoing remark.
To this end, the invention provides a compensating system for an optical transmission system including a transmitter terminal emitting a polarized optical signal, a transmission optical fiber, optional optical amplifiers and a receiver terminal, the system including first polarization dispersion compensating means including:
at least one polarization controller,
means for generating a differential group delay between two orthogonal polarization modes, the polarization controller and the differential group delay generator means being inserted between the transmission fiber and the receiver terminal, in that order, and
a control unit for the polarization controller,
and the system further including second compensating means inserted between the transmitter and receiver terminals and adapted to apply variable chromatic dispersion compensation,
in which system the control unit is adapted to control at the some time the polarization controller and the second compensating means in accordance with a measurement parameter representative of the quality of the optical signal received by the receiver terminal and in such manner as to optimize the quality.
Dynamically controlling chromatic dispersion compensation, in the same manner as the polarization controller, not only compensates chromatic dispersion in the usual sense, which is practically fixed, but also compensates the random chromatic dispersion induced by PMD.
The compensation should ideally seek to obtain the maximum improvement in signal quality, in which case the direction of the principal states of polarization e of the link as a whole between the emitted signal and the received signal coincides at all times with the direction of the polarization vector S of the received signal. In other words, the angle "PHgr" previously defined must be kept as small as possible.
However, in practice the efficacy of compensation is limited by the capabilities of current measuring, signal processing and actuator devices (polarization controller, variable chromatic dispersion compensator). Another limitation is the cost of such devices.
Another object of the invention is to propose an approach for solving this question of the rating of the compensator means according to the context of the transmission system and also the required performance.
According to this other aspect of the invention the control unit of the first and second compensating means has a response time and an accuracy such that the angle between the direction of the principal states of polarization of the link as a whole between the emitted signal and the received signal and the direction of the polarization vector of the received signal remains at all times below a given value enabling improvement of the quality relative to a transmission system with no compensation.
The maximum angle value can be determined by experiment.
Accordingly, according to one particular aspect of the invention, the control means are rated so that this angle remains below 10 degrees and preferably below 3 degrees.
According to other features of the invention, the measurement parameter is the degree of polarization of the received optical signal or the spectral width of the modulation of the electrical signal obtained by detecting the received optical signal and the control means are adopted to maximize this parameter.
According to a variant which improves the efficacy of compensation, the measuring parameter is the weighted product of the degree of polarization of the received signal by the spectral width of the modulation of the electrical signal obtained by detecting the received signal, and the control means are adapted to maximize this product.
The invention also provides an optical transmission system incorporating the compensating system defined hereinabove. The system can be a single-channel system, i.e. a system designed to convey a signal conveyed by a single carrier wavelength, or a wavelength division multiplex (WDM) system, i.e. a system designed to convey a signal made up of a plurality of channels with different carrier wavelengths. In the latter case, specific compensation must be applied to each channel. To this end the system according to the invention includes means for extracting a channel at the receiver and a compensating system associated with that channel.